Part 7 of 9 · 1 min read
Defence in depth
Many layers, so one failure is not the end.
BM argues against putting all eggs in one basket. Defence in depth means stacking protections so that if one fails, the next still holds. Its layers:
- Firewalls. They limit which traffic reaches which ports. A host firewall (such as IPTables) can be tailored to local services, with another at the perimeter. An application firewall can throttle some address ranges and spot some attacks.
- Logging. Aggregated logs do not prevent attacks, but they help detect and recover. After a patch, logs can show whether the hole was used before.
- Intrusion detection and prevention (IDS and IPS). These watch inside the perimeter for suspicious behaviour. IDS reports it and IPS can also stop it. They are heuristic, so the starting rules may be too loose or too strict. BM suggests starting with a passive IDS to tune the rules.
- Fewer privileges. From the service-account advice: narrow credentials limit the damage of any one leak (common practice generalises this as least privilege).
Check yourself: 1. What does defence in depth protect against? 2. Why start with IDS before IPS? 3. How do logs help after a breach?